Earlier this year OpenAI amazed users with an updated ChatGPT advanced voice mode, making it sound incredibly lifelike compared to the robotic voices of assistants like Alexa or Siri. This advanced voice mode responds in real-time, can handle interruptions, giggles at jokes, and even detects a speaker’s emotions. (It also sounded a lot like Scarlett Johansson during the demo.)
![OpenAI New ChatGPT Advanced Voice Mode Now You Talk](https://www.geeky-gadgets.com/wp-content/uploads/2024/07/ChatGPT-Advanced-Voice-Mode.webp)
Starting Tuesday, this advanced voice mode will begin rolling out to paid users of ChatGPT-4o. Initially, it will be available to a small group of “Plus” subscribers, with plans to make it accessible to all Plus users by fall.
Reading Suggestion: OpenAI Introduces ChatGPT-40
ChatGPT already has a basic voice mode, but this new advanced mode could transform it into a more personal virtual assistant, similar to chatting with a friend. The ease of speaking with ChatGPT’s advanced voice mode might encourage more frequent use and challenge existing virtual assistants like Apple’s Siri and Amazon’s Alexa.
However, this new feature raises some important questions:
- Will it understand users with different speech patterns?
- Will users trust this human-sounding AI assistant too much, even when it makes mistakes?
OpenAI originally planned to roll out this feature in June but delayed it by a month to ensure safety and reliability. They tested the voice capabilities with over 100 testers speaking 45 different languages across 29 regions.
To ensure safety, the voice mode will only use four pre-set voices created with voice actors to prevent impersonation. It will also block requests for music or other copyrighted audio and have the same protections as ChatGPT’s text mode to avoid generating harmful content.
Reading Suggestion: Elon Musk Launches Artificial Intelligence Company xAI: What to Know
One significant change from the May demo is the removal of the voice that sounded like Scarlett Johansson. OpenAI clarified that the voice was created with a different actor but decided to pause its use out of respect after the actor raised concerns.
This launch comes after OpenAI announced it was testing an AI-powered search engine, hinting at its growing portfolio of consumer AI tools and potential competition with Google’s search dominance.